The human pancreatic cancer cell line, BxPC3, was obtained from the American Type Culture Collection. The cells were cultured in RPMI1640 (FUJIFILM Wako Pure Chemical Corporation), supplemented with 10% heat-inactivated fetal calf serum (Capricorn Scientific), penicillin (100 U/ml) and streptomycin (100 µg/ml) (FUJIFILM Wako Pure Chemical Corporation) and incubated at 37˚C in 95% humidified air with 5% CO₂. Ibuprofen (IB) and warfarin were purchased from FUJIFILM Wako Pure Chemical Corporation. Dansylsarcosine (DNSS) was purchased from Sigma-Aldrich (Merck KGaA). All other reagents were of analytical grade.

A mixture of 4-(chloromethyl)benzyl alcohol (1.2 g; 7.66 mmol) and AgNO₃ (5.2 g; 30.6 mmol) in CH₃CN (12 ml) was stirred at 40˚C for 16 h. After cooling to room temperature, the white precipitate was removed by filtration and the filtrate was evaporated. The residue was dissolved in CH₂Cl₂, the resulting white precipitate was removed again by filtration and the filtrate was evaporated. The resulting residue of 4-(hydroxymethyl)benzyl nitrate (7.66 mmol) was dissolved in CH₂Cl₂ (35 ml), and ibuprofen (790 mg; 3.83 mmol), dicyclohexylcarbodiimide (1.2 g; 5.75 mmol) and 4-dimethylaminopyridne (46 mg; 0.38 mmol) were added to the aforementioned solution and stirred at room temperature for 17 h. After dilution with 50 ml of CH₂Cl₂, the mixture was washed with saturated aqueous NaHCO₃ and brine (saturated NaCl aqueous solution), the residual water was removed using MgSO₄ and subsequently evaporated. The residue was purified by flash column chromatography using silica gel (hexane/ethyl acetate=8/1 v/v) to give NIB (1.08 g; 76%). NIE was synthesized according to a previous report (20) (Fig. 1). ¹H and ¹³C NMR spectra were recorded on a JEOL ECA 500 spectrometer operating at room temperature. Chemical shifts were reported in parts per million (δ) relative to the residual solvent peak. Multiplicities were described as singlet (s), doublet (d), doublet of doublets (dd), triplet (t) or multiplet (m). Coupling constants (J) were reported in hertz (Hz).

To obtain the binding parameters of IB to HSA, the ultrafiltration method was used. Ultrafiltration was performed using Amicon Ultra-0.5 ml Centrifugal Filters (10 kDa cutoff; 500 µl) from Merck KGaA. After centrifugation at 5,000 x g for 30 min at 25˚C, the concentration of IB in 50 µl of filtrate, also known as the concentration of unbound drug to HSA, was measured by HPLC. The HPLC system used in the present study was a Jasco model LC-2000Plus HPLC system (JASCO Corporation). YMC-PACK ODS AM-303 (5 µm particle size; 250x5 mm internal diameter; YMC Co. Ltd.) was used as the stationary phase and was maintained at 40˚C. A total of two solvents, solvent A (50 mM sodium dihydrogen phosphate) and solvent B [50 mM sodium dihydrogen phosphate and acetonitrile (30:70, v/v)] were used as mobile phases at a flow rate of 1 ml/min. Binding parameters were obtained from the Scatchard plot using the following equation: r/[D_f]=-r x K_a + n x K_a. K_a, n, [D_f] and r were the binding affinity, the number of binding sites of IB, the concentration of unbound IB to HSA and the number of moles of IB bound per mole of HSA, respectively.

Warfarin (WF) and dansylsarcosine (DNSS) were used as site I and site II fluorescent probes on HSA, respectively (21). Fluorescence spectra of probes were measured using a Hitachi F-2500 fluorescence spectrophotometer at 25˚C (Hitachi, Ltd.). The concentrations of WF, DNSS and HSA used were 10 µM. The excitation wavelengths used for WF and DNSS were 320 and 350 nm, respectively.

NO production in aqueous solution was evaluated by measuring nitrite (NO₂^-) and nitrate (NO₃^-), the stable end-products of NO breakdown, using the NO₂/NO₃ Assay Kit-C II (Dojindo Laboratories, Inc.). Briefly, NO₂^- and NO₃^- (NOx) were measured according to the manufacturer's protocol at 1, 2, 3, 24 and 48 h after dissolving 100 µM of NIB or NIE in PBS. For the detection of NO inside cells, cells were seeded in a 6-well culture plate at 5x10⁵ cells/well and cultured overnight at 37˚C. The medium was replaced with PBS containing diaminofluorescein-FM diacetate (DAF-FM DA; 10 µM; Goryo Chemical, Inc.) and cultured at 37˚C for 1 h. After culturing, cells were washed three times with PBS, treated with NIB or NIE (200 µM) for 5 min at room temperature and imaged using a fluorescence microscope.

Live and apoptotic cell numbers were determined using the Muse^® Annexin V and Dead Cell kit (Cytek Biosciences) according to the manufacturer's instructions. Briefly, cells were seeded at a density of 2x10⁵ cells/well in 6-well plates. After 12 h, 50, 100 or 200 µM of NIE or NIB was added to each well and the plates were incubated for 6, 12 and 24 h at 37˚C. The cells were then washed twice with PBS, trypsinized (FUJIFILM Wako Pure Chemical Corporation) at 37˚C until cells were detached and mixed well with the Muse^® Annexin V and Dead Cell Assay kit reagents. Samples were measured using a Muse^® Cell Analyzer (MuseSoft; Version 1.5.0.0; Cytek Biosciences).

LDH is a cytoplasmic enzyme that it is released into the cell culture medium when the cell membrane is damaged. As the released LDH is stable in cell culture medium, it can be used as an indicator of the number of dead cells or cells with damaged cell membranes (8). Cell toxicity was evaluated using the Cytotoxicity LDH Assay Kit-WST (Dojindo Laboratories, Inc.) according to the manufacturer's protocol. Briefly, cells were seeded at a density of 3x10³ cells/well in 96-well white, flat-bottom plates. After overnight incubation at 37˚C, the cells were incubated with NIB (200 µM) or NIE (200 µM) at 37˚C for 72 h. Subsequently, the cells were incubated for 1 h at room temperature with the LDH reagent and luminescence was measured. Cytotoxicity (%) was estimated using the ratio between the concentration of LDH in the culture medium and the whole cell.

Cellular caspase 3/7 activity was measured using the Caspase-Glo 3/7 Reagent (Promega Corporation). Briefly, cells were seeded at a density of 1x10⁴ cells/well in 96-well plates. After incubation with NIB (200 µM), NIE (200 µM) or staurosporine (1 µM) for 6 h at 37˚C, Caspase-Glo 3/7 Reagent was added to each well. After incubation at room temperature for 1 h, the luminescence of each sample was measured using a plate-reading luminometer.

The mean values of each group were compared using a one-way ANOVA followed by Tukey's multiple comparison test. P<0.05 was considered to indicate a statistically significant difference. Assays were conducted in triplicate.

NIB and NIE were successfully synthesized. The chemical structure of NIB was confirmed by NMR (Fig. S1). ¹H-NMR results were as follows: (500 MHz, CHLOROFORM-D) δ 7.32 (d, J=8.0 Hz, 2H), 7.24 (d, J=8.0 Hz, 2H), 7.19 (d, J=8.0 Hz, 2H), 7.09 (d, J=8.0 Hz, 2H), 5.40 (s, 2H), 5.11 (s, 2H), 3.76 (q, J=7.3 Hz, 1H), 2.46 (d, J=7.4 Hz, 2H), 1.88-1.83 (m, 1H), 1.51 (d, J=7.4 Hz, 3H) and 0.91 (d, J=6.9 Hz, 6H). ¹³C-NMR results were as follows: (126 MHz, CHLOROFORM-D) δ 174.6, 140.8, 137.7, 137.6, 132.1, 129.5, 129.3, 128.2, 127.3, 74.5, 65.8, 45.3, 45.1, 30.3, 22.5 and 18.5.

To evaluate the binding properties of NIB and NIE to HSA, a quantitative displacement experiment was performed (Table I). Both NIB and NIE significantly decreased the binding constant of IB to HSA, but they had no significant effect on the number of binding sites. These results demonstrated that both NIB and NIE competitively inhibited the binding of IB to HSA. Fluorescence displacement experiments were performed using WF and DNSS, which are fluorescent probes for the drug binding sites I and II, respectively (Fig. 2). Although the degree of substitution was different, only DNSS was significantly substituted with both NIB and NIE. These results suggested that both NIB and NIE bind specifically to site II of HSA, as does IB.

Generally, NO donors liberate NOx by hydrolysis in aqueous solution (22). Therefore, the release of NOx from NIB and NIE in PBS was measured (Fig. 3A). Although the release of NOx from NIB was observed immediately after dissolution, no release of NOx from NIE was observed over the 48 h time period measured. Since the RPMI1640 medium used for culturing BxPC cells contains a large amount of nitrate ions, it is difficult to measure intracellular NO. Therefore, whether NIB and NIE released NO within cells was investigated. NO from NIB or NIE in BxPC3 cells was confirmed using DAF-FM DA, a cell membrane-permeable, NO-specific fluorescence probe. It was demonstrated that both NIB and NIE released NO within the cells (Fig. 3B). This result potentially suggests that a factor, such as an intracellular hydrolase, may be involved in the release of NO from NIE.

To evaluate the cytotoxicity of NIB and NIE against the BxPC3 pancreatic cancer cell line, annexin-positive cells were detected after the addition of NIB or NIE. It was demonstrated that both NIB and NIE significantly induced cell death in both a time-(Figs. 4A, S2) and concentration-dependent manner (Figs. 4B, S2), whereas IB had no significant effect on cell death. Next, an LDH assay was performed to evaluate cell membrane damage caused by NIB or NIE (Fig. 5).

To investigate the mechanism of cell death induction by NIB and NIE, the effect of the nonspecific caspase inhibitor z-VAD FMK on cell death was investigated (Fig. 6A). Cell death induced by both NIB and NIE was significantly suppressed in the presence of z-VAD FMK compared with cells not treated with z-VAD FMK. However, no significant activation of caspase 3 and 7, which serve important roles in apoptosis, was observed in cells treated with NIB and NIE at 6 and 24 h, despite treatment with staurosporine, a representative activator of caspase 3/7 (Figs. 6B and S1). These results suggested that NIB and NIE induced cell death through a non-caspase 3/7 pathway.